Physiology & Behavior - unimi.it Travai… · food (Phase 2), dogs engaged in behaviors indicating...
Transcript of Physiology & Behavior - unimi.it Travai… · food (Phase 2), dogs engaged in behaviors indicating...
�������� ����� ��
How good is this food? A study on dogs’ emotional responses to a potentiallypleasant event using infrared thermography
Tiziano Travain, Elisa Silvia Colombo, Laura Clara Grandi, EugenioHeinzl, Annalisa Pelosi, Emanuela Prato Previde, Paola Valsecchi
PII: S0031-9384(16)30105-6DOI: doi: 10.1016/j.physbeh.2016.03.019Reference: PHB 11256
To appear in: Physiology & Behavior
Received date: 7 August 2015Revised date: 15 March 2016Accepted date: 16 March 2016
Please cite this article as: Travain Tiziano, Colombo Elisa Silvia, Grandi Laura Clara,Heinzl Eugenio, Pelosi Annalisa, Previde Emanuela Prato, Valsecchi Paola, How good isthis food? A study on dogs’ emotional responses to a potentially pleasant event using in-frared thermography, Physiology & Behavior (2016), doi: 10.1016/j.physbeh.2016.03.019
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
HOW GOOD IS THIS FOOD? A STUDY ON DOGS’ EMOTIONAL RESPONSES TO A
POTENTIALLY PLEASANT EVENT USING INFRARED THERMOGRAPHY
Tiziano Travaina,*
, Elisa Silvia Colombob, Laura Clara Grandi
c, Eugenio Heinzl
d, Annalisa Pelosi
e,
Emanuela Prato Prevideb, Paola Valsecchi
a
a Dipartimento di Neuroscienze, Unità di Biologia del Comportamento, Università degli Studi di
Parma, Parco Area delle Scienze 11/A, 43124 Parma, Italy (T.T.: [email protected];
P.V.: [email protected])
b Dipartimento di Fisiopatologia Medico-Chirurgica e dei Trapianti, Sezione di Neuroscienze,
Università degli Studi di Milano, via Fratelli Cervi 93, 20090 Segrate (MI), Italy (E.S.C.:
[email protected]; E.P.P.: [email protected])
c Dipartimento di Neuroscienze, Unità di Fisiologia, Università degli Studi di Parma, via Volturno
39, 43125 Parma, Italy (L.C.G.: [email protected])
d Dipartimento di Scienze Veterinarie e Sanità Pubblica, Università degli Studi di Milano, via
Celoria 10, 20133 Milan, Italy (E.H.: [email protected])
e Dipartimento di Neuroscienze, Unità di Psicologia, Università degli Studi di Parma, Piazzale San
Francesco 1, 43121 Parma, Italy (A.P.: [email protected])
*Corresponding author.
Correspondence address: Dipartimento di Neuroscienze, via Volturno 39, 43125 Parma, Italy
E-mail address: [email protected] (T. Travain).
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Abstract
Understanding how animals express positive emotions is becoming an interesting and
promising area of research in the study of animal emotions and affective experiences.
In the present study, we used infrared thermography in combination with behavioral
measures, heart rate (HR) and heart rate variability (HRV), to investigate dogs’ emotional responses
to a potentially pleasant event: receiving palatable food from the owner.
Nineteen adult pet dogs, 8 females and 11 males, were tested and their eye temperature, HR,
HRV and behavior were recorded during a 30-minutestestconsisting of three 10-minutes
consecutive phases: Baseline (Phase 1), positive stimulation through the administration of palatable
treats (Feeding, Phase 2) and Post-feeding condition following the positive stimulation (Phase 3).
Dogs’ eye temperature and mean HR significantly increased during the positive stimulation
phase compared with both Baseline and Post-feeding phases. During the positive stimulation with
food (Phase 2), dogs engaged in behaviors indicating a positive emotional state and a high arousal,
being focused on food treats and increasing tail wagging. However, there was no evidence of an
increase in HRV during Phase 2 compared to the Phase 1, with SDNN significantly increasing only
in Phase 3, after the positive stimulation occurred.
Overall results point out that IRT may be a useful tool in assessing emotional states in dogs
in terms of arousal but fails to discriminate emotional valence, whose interpretation cannot
disregard behavioral indexes.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Highlights
1. Eye temperature and HR increase while receiving treats, no clear pattern of HRV emerges.
2. Tail wagging could be a useful indicator of positive emotional state.
3. Eye temperature can assess the arousal dimension of dogs’ emotions in a positive situation.
Keywords: Dog; Positive emotion; Behavior; Infrared thermography; Heart Rate; Heart Rate
Variability
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
1. Introduction
Considerable research showed that many animal species besides humans express emotions
through a variety of observable signals [1–7]. Emotions can be defined as psychological states
occurring when an individual is exposed to specific environmental and/or social stimuli, represent
an adaptive interface between the individual and its environment, and guide the selection of
appropriate behavioral decisions [8–10]. Psychological research on humans indicates that emotions
have a multi-component character and incorporate subjective feelings, physiological activation,
motor expressions, cognitive appraisals, and behavioral tendencies [8,11,12]. This complexity of
emotional states makes their investigation in non-human animals a challenge. According to one of
the current approaches to the study of human emotions (i.e. the dimensional perspective), emotional
states are characterized by at least two main dimensions: arousal (low to high activation) and
valence (positive to negative) [13–15]. Recently, this two-dimensional model has been applied to
the understanding of the role that emotions play in animal welfare [16,17]. Most studies focused on
emotions induced by distress and negative experiences [18–20] but a growing number of
researchers pointed out that animal welfare also entails the presence of positive emotional states
[21–23].To understand and assess animal emotions the measurement of positive and negative
valence of affective states is therefore important and an objective evaluation needs a combination of
behavioral, physiological, and cognitive markers. However, some parameters can be ambivalent and
difficult to interpret when considered separately. For example, HR can increase in both positive and
negative emotional states, since it reflects arousal rather than valence [21,24,25]. The beat – to –
beat variation (Heart Rate Variability – HRV), that reflects changes in the activity of the autonomic
nervous system [26], is not differentially influenced by the valence of the test situations such as
accessing to popcorn (positive situation) and crossing a black ramp (negative situation) in pigs [27].
Even behavioral measures are not always easy to interpret [15]: lip licking in dogs has previously
been explained as a signal of stress [28,29] but has also been related to an increased arousal
determined by the reunion with the owner after a long period of separation [30] and some
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
researchers have considered it as an appeasement signal [31]. Self-grooming in dogs can be
considered as an index of relaxation and appropriate self-maintenance but can also be associated
with an attempt to relieve stress or anxiety [32].
While the literature on positive emotions in farm animals has grown [5,27,33–41], the same
topic has received little attention in companion animals [42–47]. Dogs’ positive affective states
have been investigated in female laboratory beagles tested in four different experimental protocols:
Burman and colleagues [42] used the ‘cognitive bias test’ with a food treat in a bowl; Rehn and
colleagues [45] used a separation/reunion to a familiar person paradigm; McGowan and colleagues
[43] used a problem solving operant task comparing different rewards: food, human or dog contact;
Zupan and colleagues [47] used more/less attractive food/social stimuli testing dog in a runaway
motivation test. Kuhne and colleagues [44] evaluated pet dogs’ emotional state and behavioral
responses to physical human–dog contact by a familiar or unfamiliar person. In a pilot study, Gygax
and colleagues [46] conducted a test to assess behavioral and neural indicators (fNIRS) of positive
emotional states in dogs using three types of human interactions: verbal, physical or both.
Overall, these studies provided interesting evidence that certain circumstances elicit positive
emotional states in dogs that can be measured through behavioral (e.g. tail wagging, proximity and
contact seeking, gazing, stress signals, and vocalizations [42–47]), physiological (HR and HRV
[44,47]), endocrine (oxytocin and cortisol [45]), and neural (hemodynamic [46]) indicators.
The aim of the present study was to introduce another physiological parameter, i.e.
superficial body temperature, in combination with behavioral, heart rate and heart rate variability
measurements, to investigate positive emotions in a sample of pet dogs receiving food treats from
their owner. Superficial temperature was measured using infrared thermography (IRT), a remote
and non-invasive technique that detects changes in peripheral blood flow. This technique has been
recently used to explore physiological correlates of stress and emotions in animals [36,48–56] with
only one study on dogs [56],whereas heart rate and heart rate variability are regarded as suitable
tools to investigate the role of ANS in the modulation of affect and emotion [17,20,35,55,57,58].
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Given that very palatable food is considered a positive reward [15,21,36,38,40,59] and based
on the recent evidence by Zupan and colleagues [47], we hypothesized that a very palatable food
treat received from the owner should attract dogs’ attention toward her/him, significantly increasing
gazing behavior [60] and should also be a source of arousal, determining an increase in HR
[47,61,62]. In addition, since HRV reflects the continuous interplay between the two branches of
the ANS and it has been suggested that HRV changes are associated to the positive valence of
emotional state [34,35,47], we expected that heart rate variability parameters should vary. HRV
could increase [63] or decrease [47] and therefore at the present stage of knowledge on dog
emotional reaction is difficult to make a more precise hypotheses. Finally, if receiving treats
induced in dogs a positive emotional state we should observe an increase in tail wagging [43].
Making predictions on eye temperature changes is difficult, given that the available evidence
is relative to a variety of body surface areas, tools and species tested with opposite results (comb in
hens: [36]; nose in cows: [41]; eye in dogs: [56]; eye in horses: [54]; nose in macaques: [48,52,64]).
However, in dogs eye temperature correlates with core body temperature [56] and thus it is possible
that eliciting an attentional state would result in an increase in eye temperature due to a general state
of arousal. To test these predictions, dogs’ eye temperature, heart rate, heart rate variability and
behavior were recorded prior (Phase 1), during (Phase 2) and after (Phase 3) food treats delivery.
2. Materials and methods
2.1 Subjects
The subjects were 19 healthy dogs (8 females, 11 males) of different breeds and body size,
whose ages ranged from 2 to 11 years (mean = 6.36 years, SD = 2.72 years). The sample included
13 pure-breed dogs (1 Jack Russell Terrier, 2 Australian Kelpies, 1 Border Collie, 1 Irish Setter, 1
Irish Red and White Setter, 4 Labrador Retrievers, 2 Golden Retrievers, 1 Newfoundland) and 6
mixed-breed dogs (1 miniature size, 2 small size, 3 medium size). All the dogs were kept for
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
companionship, lived within the human household, were accustomed to share daily activities with
their owner (e.g. travel by car, going to unfamiliar places, encountering unfamiliar humans), and
were used to wearing a harness on daily walking.
2.2 Procedure
The study was conducted at the Canis sapiens Lab of the University of Milan (Italy). On
arrival, the human-dog pairs were escorted to a waiting room where the procedure was briefly
described to the owners who were asked to provide their written consent to record behavior and to
use the collected data, according to the national Privacy Law 675/96. To apply the heart rate
monitor (Polar), the dogs were sheared under the right and left armpits for a surface of
approximately 10 cm2each side. Then the Polar was fixed to the dog chest by means of a belt (see
data collection paragraph for further details). After this manipulation, dogs were allowed to freely
explore the waiting room for an additional period of 5 minutes to relax and to familiarize with the
video camera operator and the infrared thermography technician.
The test took place in an unfamiliar adjacent bare room (3.00 x 5.00 m) equipped with one
chair, a carpet and a video camera (Leica Dicomar, Panasonic, Japan) mounted on a tripod. During
the test the video camera operator (E.S.C.), the infrared thermography technician (E.H.) and the
owner were present. To minimize extraneous noise and disturbance, testing was conducted on
weekends over a period of four months. The testing environment was air-conditioned and thus
temperature (22 °C) and humidity (40%) remained constant during the procedure[56].Owners were
asked not to feed their dogs for at least four hours prior testing.
The test procedure consisted of three consecutive phases:
Phase 1, Baseline: After the dog and the owner entered the experimental room, the Polar was
switched on and the owner was asked to sit on the chair while the dog, on a 2 m long leash, could
remain close to her/him or move around for 10 minutes. The aim of this phase was to obtain
baseline values for each dog, consequently only data from the last 5 minutes were used for
statistical analysis (see Statistical Analysis and Results sections for the rationale of this choice).
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Phase 2, Feeding (i.e. positive stimulation): The owner remained seated and attracted the
dog’s attention by showing the dog treats in her/his hands and gave the dog treats at approximately
20 sec. intervals for a 10-minute period. Treats consisted of 1-gram chicken croquettes (Nature
Snack - Mini sandwiches for dog, Ferribiella, Italy). Owner compliance to our instructions resulted
in some variation in the number of treats given to the dogs (mean = 30.90, SD = 13.47) nevertheless
the number of croquettes received was not correlated with any physiological parameter measured.
Phase 3, Post-feeding: As in Phase 1, the owners remained seated while the dog, on leash,
could remain close to her/him or move around for 10 minutes.
2.3Behavioral measurements
Behavioral data were scored from videos using Solomon Coder beta® 15.01.13 (ELTE
TTK, Hungary). The ethogram consisted of 13 non-mutually exclusive behaviors and of 2
behavioral categories, namely Dynamic (walking, jumping and jumping on the owner) and Static
(standing still, sitting, or lying down). These two behavioral categories were mutually exclusive but
could co-occur with the other 13 behaviors, thus a dog could sit in front of the owner and gaze at
her/his face (static + gazing owner) or could walk while exploring the environment (dynamic +
attention). Behaviors and categories are detailed in Table 1 and were recorded either as duration or
as frequencies.
Only behavioral patterns that were displayed with a mean frequency>1 event/min and with a
mean duration >1% of the time in at least one of the three experimental phases were included in the
statistical analysis [65]. Moreover, those behaviors exhibited only in one of the three experimental
phases were not statistically analyzed.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Table 1. Description and measure of coded behaviors and behavioral categories recorded during the
test.
Behavior Description Frequency/Duration Interobserver reliability
Dynamic
Walking, jumping and jumping on the
owner
D (% on total time) r = 0.880, P < 0.001
Static Standing still, sitting, or lying down D (% on total time) r = 0.604, P = 0.008
Shaking off Rapid movements of body shaking F (events/min) r = 1.000, P = 0.000
Yawning
Deeply inhale through wide open
mouth
F (events/min) r = 0.921, P < 0.001
Nose/lip licking
Rapid extension and flicking of the
tongue on the nose or between the lips
F (events/min) r = 0.725, P = 0.001
Panting Breathing with short, quick breaths D (% on total time) r = 0.720, P = 0.001
Avoidance
Lateral movement of the head to avoid
being focused by thermographic
camera
F (events/min) r = 1.000, P = 0.000
Rest Resting, laying on the ground D (% on total time) r = 0.986, P < 0.001
Gazing experimenter Looking at the experimenter D (% on total time) r = 0.922, P < 0.001
Gazing owner Looking at the owner D (% on total time) r = 0.977, P < 0.001
Gazing owner’s
hand/food
Looking at the owner’s hand or at the
food held by the owner
D (% on total time) r = 0.976, P < 0.001
Hand sniffing/licking
Sniffing or licking owner’s hand,
regardless of the presence or absence
of the food
D (% on total time) r = 0.835, P < 0.001
Touching with paw Touching the owner with the paw D (% on total time) r = 0.904, P < 0.001
Attention Visual/olfactory exploration of the
environment
D (% on total time) r = 0.920, P < 0.001
Tail wagging Tail hanged in a relaxed manner at
half‐mast[66] and wagged D (% on total time) r = 0.600, P = 0.008
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
To evaluate whether the thermographic camera directed towards the dogs’ muzzle could be
perceived as stressing (see [56]), during coding, a marker indicating the presence of the
thermographic camera accompanied each behavioral element. The dogs’ behavior was coded from
videos by one of the authors (T.T.) whereas a second independent coder analyzed 6 dogs (31.58%
of the subject sample) to assess interobserver reliability.
2.4 Physiological measurements
The thermographic infrared images were captured by a certified technician (E.H.) using a
portable IRT camera (AVIO TVS500® camera, NEC, Japan) with standard optic system, and
analyzed with IRTAnalyzer Software® (Grayess, FL, USA). To calibrate the camera reflectivity
temperature, samples were taken and emissivity was set at 0.97. Several images per dog were
collected, to select those images that provided the most optimal operating conditions for analysis
(90° angle and 1 m of distance). During the whole study 982 (per dog: mean = 51.68, SD = 10.83;
minimum = 31; maximum = 77) images were analyzed to evaluate the emission of eyes lachrymal
sites. The maximum temperature for each lachrymal site was determined using an Instantaneous
Field of View of 1.68 mm at 1 m of distance, within an oval area traced around the eye, including
the eyeball and approximately 1 cm surrounding the outside of the eyelids. Lacrimal caruncle was
chosen as target area on the basis of information derived from previous studies [67,68] and because
its temperature is not influenced by the presence of hair. To optimize the accuracy of the
thermographic image and to reduce sources of artefacts, before every testing session the same
image of a Lambert surface was taken to define the radiance emission and to nullify the effect of
surface reflections on tested animals [69]. Moreover, the images were corrected for environmental
temperature. Only images perfectly on focus were used (Figure 1). To determine the caruncle’s
temperature, the maximum temperature within a circular area traced around the target area was
measured. This maximum value was used for subsequent analysis.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Fig. 1. Thermographic image (photo 1) and corresponding picture (photo 2) of Easy, Staffordshire
bullterrier mix, during Phase 2 (Feeding). A! is the lacrimal caruncle and the hottest spot on the eye.
B! is the second hottest spot on the eye and it is highlighted for control purposes. 1 is the marker for
the oval area traced around the eye.
Heart rate data were collected using a Polar® RS800CX human HR monitor (Polar®
Electro, Finland). The Polar WearLink® strap was positioned around the dog thorax and the size
was adjusted to provide a tight but comfortable fit. Farmacare ultrasound transmission gel
(Farmacare, Italy) was applied to the 2 electrodes of the Polar WearLink® strap. The electrodes
were positioned over the right and left axillary regions. The Polar® watch computer was fixed
dorsally to the WearLink strap.
The Polar® was set on the R-R interval-recording mode and data collection lasted for the
whole duration of the experiment. R-R interval data were analyzed using Kubios HRV software
(Version 2.1 Biosignal Analysis and Medical Imaging Group, Department of Applied Physics,
University of Eastern Finland, Kupio, Finland). Prior to analyses, artifacts were removed using
Kubios’ inbuilt artifact correction feature. The artifact tolerated was not more than 1% on the total
length of the recording, for each dog. Heart rate (HR, beats per minute) and HRV parameters were
calculated for each experimental phase. The following time-domain variables were chosen for
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
analysis: mean HR (bpm), root mean square of the standard deviation (RMSSD, ms), standard
deviation of R-R intervals (SDNN, ms) and the ratio SDNN/RMSSD.
2.5 Statistical analysis
Interobserver reliability was assessed using Spearman correlations analyzing all 3 phases of
6 dogs (n = 18) and it was significant for all the behaviors recorded with r ranging from 0.6 to 1
(Table 1).
To evaluate any habituation effect that could have occurred during Phase 1 (Baseline), a
preliminary analysis was carried out to detect differences between the first and the last 5 minutes of
the Phase 1. Paired T-tests were carried out for all physiological parameters and for the most
relevant behaviors (i.e. Dynamic and Rest). According to the outcome of the Paired T-tests, detailed
in Results section, the last 5 minutes of the Baseline were used in all the subsequent analysis.
Eye temperature was analyzed using for each dog all the thermographic images taken during
each phase and dog’s sex as factor.
Behavioral and physiological data were entered in a Generalized Linear Mixed – effect
Models Analysis (GLMM): for eye temperature, we considered the random effects of phase nested
within dog identity, whereas behaviors and cardiac data included dog identity as the random effect.
Separate models were set up for each dependent variable measure (behaviors, eye temperature, HR
and HRV). Residual funnel graph confirmed the model residuals homoscedasticity and the Shapiro
test established that they did not significantly diverge from normality. All models included the
explanatory variables phase and sex; moreover, Dynamic behavior was also used as co-variate
variable for HR and HRV parameters. All models were set up with a step-wise forward procedure:
starting from a null random model, predictors (principal effects and then interaction term) were
added only if they determined a reliable change into the model fitting (Akaike Information Criterion
– AIC – and Schwarz’s Bayesian Information Criterion – BIC). Planned contrasts with Bonferroni
correction were used for paired comparisons.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
All statistical analyses were carried out using R 3.1.3 (R development Core Team) packages:
car – Companion of Applied Regression [70], lme4 – Linear Mixed-Effects Models using Eigen and
S4 [71] and lmerTest [72].
3. Results
3.1 Preliminary analysis
Between the first and the last 5 minutes of the Phase 1 (Baseline), HR values significantly
decreased (bpm 100.75 14.83 versus 93.64 14.64; t18 = 2.94, P = 0.004) while HRV increased
(SDNN: ms 91.86 29.92 versus 115.51 40.69; t18 = -4.08, P = 0.001; RMSSD: ms 67.53 28.85
versus 81.08 26.09; t18 = -2.35, P = 0.031; SDNN/RMSSD: 0.77 0.33 versus 0.76 0.32; not
significant). Eye temperature did not vary (°C 36.73 0.47 versus 36.74 0.46; t18 = 0.27, P =
0.794). Besides, while Dynamic behavior did not change, there was an increment in resting time
(Dynamic: % 7.07 4.76 versus 4.87 5.32; t18 = 1.74, P = 0.099; Resting: % 11.65 20.35 versus
28.28 30.03; t18 = -3.68, P = 0.002).
The overall pattern of changes suggested that a habituation effect to the novel environment
have occurred in the second part of Phase 1. Consequently, for all behavioral and physiological
parameters only data from the last five minutes of Phase 1 were used to analyze the difference
among Phase 1, Phase 2 and Phase 3.
3.2 Behavioral data
Dogs’ behavioral response changed across the three phases, in particular, Hand
sniffing/licking and Tail wagging significantly increased between Phase 1 and 2 and decreased
between Phase 2 and 3 returning to the baseline values (Table 2). Conversely, Rest, Gaze
experimenter and Attention decreased significantly between Phase 1 and 2, whereas they increased
between Phase 2 and 3. Gazing owner significantly varied among the three Phases, but not
significant pairwise comparisons emerged.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Panting did not change across the experimental phases whereas sex difference occurred
(Table 2): overall, males panted for a greater percentage of time than females and no differences
among phases within each sex emerged. Some behaviors were expressed only in Phase 2 as
touching the owner with the paw (1.37% of the time) and gazing owner’s hand/food (59.13% of the
time).
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Table 2. Mean, SD, and statistical results of expressed behaviors.
Behavior
Phase 1
Baseline
mean ± SD
Phase 2
Feeding
mean ± SD
Phase 3
Post-feeding
mean ± SD
Principal effects and interactions (pair-wise comparisons)
Dynamic 1 4.87 ± 5.32 2.82 ± 4.06 3.03 ± 3.32
Behavior not analyzed because it was used as a covariate for heart rate
parameters analysis
Static 1 95.13 ± 5.32 97.18 ± 4.06 96.97 ± 3.32 Phase: F2,36= 1.53, P = 0.23
Shake off 2,3
0.04 ± 0.09 0.00 ± 0.00 0.03 ± 0.06 Not analyzed
Yawning 2,3
0.07 ± 0.10 0.01 ± 0.02 0.07 ± 0.09 Not analyzed
Nose/lip licking 2 1.29 ± 1.44 0.91 ± 0.97 0.61 ± 0.54 Phase: F2,36 = 2.06, P = 0.014 (non-significant pair-wise comparisons)
Panting 1
♀ 0.71 ± 2.02
♂ 24.39 ± 28.94
♀ 0.29 ± 0.81
♂ 7.33 ± 13.79
♀ 5.62 ± 11.23
♂ 26.52 ± 34.36
Phase: F2,34 = 2.23, P = 0.42. Sex: F1,17 = 4.48, P = 0.019;
Phase * Sex: F 2,34 = 2.65, P = 0.56 (non-significant pair-wise
comparisons)
Avoidance 2,3
0.01 ± 0.04 0.00 ± 0.00 0.00 ± 0.00 Not analyzed
Rest 1 28.28 ± 30.03 0.25 ± 0.98 30.53 ± 26.99 Phase: F2,36 = 16.11, P < 0.001 (
a P <0.001
b P = 0.768;
c P <0.001)
Gaze owner 1 9.27 ± 9.98 5.29 ± 5.80 16.82 ± 19.53 Phase: F2,36 = 5.61, P = 0.008 (non-significant pair-wise comparisons)
Gaze owner’s
hand/food 1,3
0.00 ± 0.00 59.13 ± 20.97 0.00 ± 0.00 Not analyzed
Gaze experimenter 1 12.03 ± 8.51 2.87 ± 3.36 16.99 ± 12.62 Phase: F2,36 = 16.83, P < 0.001(
a P = 0.006;
b P = 0.9;
c P <0.001)
Hand sniffing/licking 1 0.47 ± 0.98 20.72 ± 17.19 1.30 ± 2.33 Phase: F2,36= 27.15, P < 0.001 (
a P < 0.001;
b P = 0.8;
c P < 0.001)
Touch with paw 1,3
0.00 ± 0.00 1.37 ±2.21 0.00 ± 0.00 Not analyzed
Attention 1 48.77 ±22.91 11.74 ±9.78 34.35 ±17.26 Phase: F2,36 = 28.86, P < 0.001 (
a P <0.001;
b P = 0.014;
c P=0.001)
Tail wagging 1 2.02 ±3.90 27.92 ±41.70 6.20 ±11.28 Phase: F2,36 = 5.89, P = 0.006 (
a P =0.007;
b P = 0.61;
c P = 0.019)
1 Behaviors measured as duration (% of total time).
2 Behaviors measured as frequency (events/min).
3 Behaviors not analyzed because they occurred in only one phase or were rarely
displayed. a: Baseline vs Feeding;
b: Baseline vs Post-Feeding;
c: Feeding vs Post Feeding
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
3.3Physiological data
Eye temperature. Eye temperature significantly increased from Phase 1 to Phase 2 and
decreased from Phase 2 to Phase 3 (F2,693.7 = 57.62, P < 0.001, Figure 2).
Fig. 2. Boxplot representing eye temperature (°C) exhibited by the dogs during the three phases of the
experiment (1. Baseline, 2. Feeding, 3. Post-feeding). Pair-wise comparisons: *** P = 0.001.
HR and HRV. There was not significant relationship between physical activity (Dynamic) and
any of the analyzed cardiac parameters (HR: F1,49.5 = 3.37, P = 0.081; SDNN: F1,52.1 = 0.487, P = 0.488;
RMSSD: F1,49.9 = 3.24, P = 0.091; SDNN/RMSSD F1,54.2 = 2.38, P = 0.123).
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Once controlled for the effect of Dynamic, HR and SDNN changed across phases (HR: F2,35.5 =
7.59, P = 0.002; SDNN: F2,37.9 = 3.31, P = 0.04). HR increased significantly from Phase 1 to 2 and
diminished from Phase 2 to 3 returning to the baseline values. SDNN significantly increased from
Phase 2 to 3, whereas no differences occurred between Phase 1 and 3 (Figure 3).On the contrary,
RMSSD and SDNN/RMSSDdid not significantly differamong phases (RMSSD: F2,37.9 = 0.138, P =
0.871; SDNN/RMSSD: F2,38.9 = 0.829, P = 0.443; Figure 3).
Fig. 3. Boxplot representing values of heart rate (bpm), RMSSD (ms), SDNN (ms), and
SDNN/RMSSD exhibited by the dogs during the three phases of the experiment (1. Baseline, 2.
Feeding, 3. Post-feeding). Pair-wise comparisons: * P = 0.05, ** P = 0.01.
4. Discussion
The aim of the present study was to investigate eye temperature, detected through IRT, as a
potential physiological indicator of emotional states in pet dogs. In addition to eye temperature, cardiac
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
activity (i.e. HR and HRV) was monitored to understand the interplay between the sympathetic and the
parasympathetic branch of the ANS when a food stimulus was provided. The analysis of dogs’
behavior was used to provide an integrative measure of the response to the stimulation.
Mean eye temperature increased significantly during Phase 2 (Feeding), compared with the
mean values of both Phase 1 (Baseline) and 3 (Post-feeding). This result supports our prediction that
the arousal determined by the presence of food in the owner hands would have resulted in an increment
of eye temperature. The available literature concerning the use of IRT in the study of animal emotions
points out that surface body temperature can increase/decrease depending on the investigated species,
the anatomical areas and the type of stimulus used. Nasal temperature decreases in macaques during
negative emotional states [48,52,64], and in cows when exposed to a positive stimulus [41]. In hens
there is a significant drop in comb surface temperature in response to anticipation and consumption of a
palatable food reward [36] but also during exposure to unpleasant events [73]. This contrasting effect
on body temperature can be explained in terms of activation of the sympathetic branch of the ANS
which induces an increase in core temperature, reflected in the eye, and a decrease in more peripheral
body area, such as nose, face and ears, due to vasoconstriction [49,50,74,75].
Interestingly, we found that in dogs eye temperature increases both when they receive a food
treat (positive valence) and when they are stressed by a veterinary visit (negative valence) [56].In both
studies eye temperature showed the same pattern of temporal changes, even though different ranges in
the eye temperature emerged possibly due to inter-individual differences. Overall, these evidences
suggest that body surface temperature is a good index of a general state of arousal but, at this stage of
our knowledge on dogs, it does not allow the discrimination of the positive or negative emotional
valence of the stimulus itself.
The analysis of behavior provided insight that dogs perceived the present situation as positive;
in fact they did not show fear and only one dog performed avoidance of the IRT camera during Phase 1
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
(Baseline),while during the veterinary visit this was an important issue [56]. During Phase 2 (Feeding),
most of the time dogs remained oriented towards their owner’s hands, looking at the food and showing
an increased duration of tail wagging. McGowan and colleagues [43] showed that dogs facing a
problem-solving task under their control had a more intense emotional response, expressed as
frequency of tail wags, when the reward was food rather than contact with a conspecific and suggested
that tail wagging can be used as an indicator of a positive affective state in dogs. Quaranta and
colleagues [76] showed that dogs wagged their tail in an asymmetric manner in response to different
emotional stimuli and that the amplitudes of tail wagging also depend on the stimuli, being highest
when seeing the owner and lowest when seeing a cat. Being a context dependent behavior, tail wagging
could be an interesting index to assess emotional states in dogs, but breed (in our sample the most
wagging dogs were 3 Labrador retriever and 1 Golden retriever), temperament and motivation are all
factors that may affect its expression and thus further studies would be beneficial. Nose/lip licking was
rarely performed and it occurred mainly during Phase 1 when dogs entered the unfamiliar room for the
test. Despite being a signal occurring in different contexts [28–31,46], in the present study it could
indicate an initial mild discomfort in the new environment. Other behaviors signaling stress were
almost completely absent, confirming that the procedure was not stressful as was instead the veterinary
visit of the previous study, when dogs showed stress/fear signals in all the three phases of the
experiment. In conclusion, there are emerging evidences that behavioral changes might provide useful
and complementary information on the emotional valence of a given situation [5,77] and also on the
changes in the level of arousal over time due to a habituation processes to emotional events [46].
Once removed the variance due to the dynamic component of behaviors, we found a significant
increase in mean HR during Phase 2. In line with previous literature on humans and many other animal
species [24,47,63,78,79], the results of this study indicated that mean HR (corrected for dynamic
components of behavior) was linked to emotional arousal. Nevertheless, HR provides limited
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
information for accurately assessing sympathovagal regulation [80], whereas HRV analysis allows a
much more accurate and detailed determination of the functional regulatory characteristics of the ANS.
Indeed, psychological states may have an impact on sympathovagal balance in the absence of any
palpable changes in heart and/or respiration rates in a specific condition [81–83].
Recent evidence suggested that positive emotional states in animals are reflected in changes of
HRV parameters, which may provide information on valence dimension of emotions[21,35,47,58].
However the direction of changes in HRV is still controversial: Reefman and colleagues [35] suggested
that in sheep lower HR and increased RMSSD are linked to a positive emotional situation. Conversely,
Zupan and colleagues [47] hypothesized that a decrease in RMSSD may reflect a more positive
emotional state. Considering that SDNN reflects the activity of both sympathetic and parasympathetic
branches while the RMSSD reflects the contribution of only the vagal tone to the autonomic balance,
we expected a variation of HRV during the Feeding phase. Even though in Phase 2 dogs showed
behavioral signs of being positively stimulated (i.e. increase in tail wagging), HRV parameters did not
change. In phase 3, when dogs did not receive any more food, we observed a significant increment of
SDNN, however, RMSSD and SDNN/RMSSD ratio did not change. Overall results suggested that the
balance of autonomic system remained stable across the experimental phases.
Taken together these results indicated that receiving food treats might not be sufficient per se to
induce a long-term change in adult pet dogs’ autonomic system and this could be due to a number of
reasons: on one hand, differently from the two previous studies using food as a reward[43,47], we
tested adult pet dogs living within the household and accustomed to receive palatable treats from the
owner; this might have limited the positive valence of the food stimulus. On the other hand, in our
study food was provided directly to the dogs, i.e. without performing a specific action to get it, and this
might have reduced the hedonic value of the situation. In the current study dogs were tested with their
owners in order to reproduce a more suitable environmental situation for pet dogs; however, this may
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
have introduced a variability due to the owner-dog relationship that could not be controlled for. It is
important to underline how experiments with both laboratory and pet dogs are useful to provide
information on different responses to emotional stimuli; however, while the former allows the control
of more dogs’ variables (e.g., age, previous experiences, housing, food, and daily routine), the latter
allows to investigate these issues in a variety of situations close to the daily experience of a dog living
in a human household. Both these different approaches are promising, but given the limited number of
studies conducted so far, further researches are required to better identify useful behavioral and
physiological markers on dogs’ positive affective state.
5. Conclusions
Our study is the first assessing eye temperature as a measure of positive emotional states in
dogs. Results showed that IRT is suitable to detect in a non-invasive way a state of arousal but not to
assess the hedonic values of a positive stimulus, unless combined with behavioral indicators. Thus, the
validation of behavioral pattern linked to positive and negative experiences is crucial for understanding
dogs’ emotional states. Although, the presence of differences in tail wagging between phases indicated
that dogs were in a positive state, the analysis of HRV parameters gave ambiguous results suggesting
that the administration of palatable food per se might not be the sufficient to elicit an emotion with a
positive valence.
Acknowledgements
The authors thank all the dog owners who kindly participated in this study. Tiziano Travain was
supported by a doctoral grant from the Università degli Studi di Parma. Elisa Silvia Colombo was
supported by a doctoral grant from the Università degli Studi di Milano. Partial funding from
Università degli Studi di Parma (FIL 2014) to Paola Valsecchi was used. The funding sources were not
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
involved in the study design; in the collection, analysis and interpretation of data; in the writing of the
report; and in the decision to submit the article for publication.
Ethic Statement
No special permission for the use of animals (dogs) in non-invasive observational studies is
required in Italy. The relevant ethical committee is the Ethical Committee of the Università degli Studi
di Milano. All dog owners were informed about the nature and scope of the study and their written
consent was obtained before the study was initiated.
Conflict of interest
None of the authors of this paper has a financial, personal or other relationship with other
people or organizations within three years of beginning this work that could inappropriately influence
or bias the content of the paper.
References
[1] C. Darwin, The expression of emotions in man and animals, Oxford University Press, Oxford,
1872.
[2] M.W. Fox, A Comparative Study of the Development of Facial Expressions in Canids; Wolf,
Coyote and Foxes, Behaviour. 36 (1970) 49–73. doi:10.1163/156853970X00042.
[3] R.M. Seyfarth, D.L. Cheney, Meaning and emotional in animal vocalizations, Ann. New York
Acad. Sci. 1000 (2003) 32–55.
[4] P. Pongrácz, C. Molnár, Á. Miklósi, Acoustic parameters of dog barks carry emotional
information for humans, Appl. Anim. Behav. Sci. 100 (2006) 228–240.
doi:10.1016/j.applanim.2005.12.004.
[5] N. Reefmann, F. Bütikofer Kaszàs, B. Wechsler, L. Gygax, Ear and tail postures as indicators of
emotional valence in sheep, Appl. Anim. Behav. Sci. 118 (2009) 199–207.
doi:10.1016/j.applanim.2009.02.013.
[6] E. Palagi, G. Mancini, Playing with the face: playful facial “chattering” and signal modulation in
a monkey species (Theropithecus gelada), J. Comp. Psychol. 125 (2011) 11–21.
doi:10.1037/a0020869.
[7] C. Kemp, G. Kaplan, Facial expressions in common marmosets (Callithrix jacchus) and their use
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
by conspecifics, Anim. Cogn. 16 (2013) 773–788. doi:10.1007/s10071-013-0611-5.
[8] K.R. Scherer, On the Nature and Function of Emotion: A Component Process Approach, in: K.
Scherer, P. Ekman (Eds.), Approaches to Emot., Psychology Press, New York, 1984: pp. 293 –
317.
[9] R.S. Lazarus, Emotion and adaptation, Oxford University Press, New York, 1991.
[10] M.S. Dawkins, The science of animal suffering, Ethology. 114 (2008) 937–945.
doi:10.1111/j.1439-0310.2008.01557.x.
[11] K.R. Scherer, What are emotions? And how can they be measured?, Soc. Sci. Inf. 44 (2005)
695–729. doi:10.1177/0539018405058216.
[12] P.M. Niedenthal, S. Krauth-Gruber, F. Ric, Psychology of Emotion: Interpersonal, Experiential
and Cognitive Approaches, Psychology Press, New York, 2006.
[13] J.A. Russell, A circumplex model of affect., J. Pers. Soc. Psychol. 39 (1980) 1161–1178.
doi:10.1037/h0077714.
[14] L. Feldman Barrett, J.A. Russell, The Structure of Current Affect: Controversies and Emerging
Consensus, Curr. Dir. Psychol. Sci. 8 (1999) 10–14. doi:10.1111/1467-8721.00003.
[15] M. Mendl, O.H.P. Burman, E.S. Paul, An integrative and functional framework for the study of
animal emotion and mood., Proc. Biol. Sci. 277 (2010) 2895–2904. doi:10.1098/rspb.2010.0303.
[16] L. Désiré, A. Boissy, I. Veissier, Emotions in farm animals: A new approach to animal welfare
in applied ethology, Behav. Processes. 60 (2002) 165–180. doi:10.1016/S0376-6357(02)00081-
5.
[17] A. Boissy, C. Arnould, E. Chaillou, V. Colson, L. Désiré, C. Duvaux-Ponter, et al., Emotion and
cognition: a new strategy to achieve animal welfare, Prod. Anim. 20 (2007) 17–21.
[18] E.O. Johnson, T.C. Kamilaris, G.P. Chrousos, P.W. Gold, Mechanisms of stress: A dynamic
overview of hormonal and behavioral homeostasis, Neurosci. Biobehav. Rev. 16 (1992) 115–
130. doi:10.1016/S0149-7634(05)80175-7.
[19] I. Veissier, A. Boissy, Stress and welfare: Two complementary concepts that are intrinsically
related to the animal’s point of view, Physiol. Behav. 92 (2007) 429–433.
doi:10.1016/j.physbeh.2006.11.008.
[20] E. von Borell, J. Langbein, G. Després, S. Hansen, C. Leterrier, J. Marchant-Forde, et al., Heart
rate variability as a measure of autonomic regulation of cardiac activity for assessing stress and
welfare in farm animals - A review, Physiol. Behav. 92 (2007) 293–316.
doi:10.1016/j.physbeh.2007.01.007.
[21] A. Boissy, G. Manteuffel, M.B. Jensen, R.O. Moe, B. Spruijt, L.J. Keeling, et al., Assessment of
positive emotions in animals to improve their welfare, Physiol. Behav. 92 (2007) 375–397.
doi:10.1016/j.physbeh.2007.02.003.
[22] J.W. Yeates, D.C.J. Main, Assessment of positive welfare: A review, Vet. J. 175 (2008) 293–
300. doi:10.1016/j.tvjl.2007.05.009.
[23] M. Coulon, R. Nowak, J. Peyrat, H. Chandèze, A. Boissy, X. Boivin, Do Lambs Perceive
Regular Human Stroking as Pleasant? Behavior and Heart Rate Variability Analyses, PLoS One.
10 (2015) e0118617. doi:10.1371/journal.pone.0118617.
[24] C. Palestrini, E. Prato-Previde, C. Spiezio, M. Verga, Heart rate and behavioural responses of
dogs in the Ainsworth’s Strange Situation: A pilot study, Appl. Anim. Behav. Sci. 94 (2005) 75–
88. doi:10.1016/j.applanim.2005.02.005.
[25] M.T. Mendl, O.H.P. Burman, R.M.A. Parker, E.S. Paul, Cognitive bias as an indicator of animal
emotion and welfare: Emerging evidence and underlying mechanisms, Appl. Anim. Behav. Sci.
118 (2009) 161–181. doi:10.1016/j.applanim.2009.02.023.
[26] Task Force of the European Society of Cardiology, The North American Society of Pacing and
Electrophysiology, Heart Rate Variability: Standards of Measurement, Physiological
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Interpretation, and Clinical Use, Eur. Heart J. 17 (1996) 354–381.
doi:10.1161/01.CIR.93.5.1043.
[27] S. Imfeld-Mueller, L. Van Wezemael, M. Stauffacher, L. Gygax, E. Hillmann, Do pigs
distinguish between situations of different emotional valences during anticipation?, Appl. Anim.
Behav. Sci. 131 (2011) 86–93. doi:10.1016/j.applanim.2011.02.009.
[28] B. Beerda, M.B.H. Schilder, J.A.R.A.M. Van Hooff, H.W. De Vries, Manifestations of chronic
and acute stress in dogs, Appl. Anim. Behav. Sci. 52 (1997) 307–319. doi:10.1016/S0168-
1591(96)01131-8.
[29] B. Beerda, M.B.H. Schilder, J.A.R.A.M. Van Hooff, H.W. De Vries, J.A. Mol, Behavioural and
hormonal indicators of enduring environmental stress in dogs, Anim. Welf. 9 (2000) 49–62.
[30] T. Rehn, L.J. Keeling, The effect of time left alone at home on dog welfare, Appl. Anim. Behav.
Sci. 129 (2011) 129–135. doi:10.1016/j.applanim.2010.11.015.
[31] F. Kuhne, J.C. Hößler, R. Struwe, Effects of human-dog familiarity on dogs’ behavioural
responses to petting, Appl. Anim. Behav. Sci. 142 (2012) 176–181.
doi:10.1016/j.applanim.2012.10.003.
[32] N. Rooney, S. Gaines, E. Hiby, A practitioner’s guide to working dog welfare, J. Vet. Behav.
Clin. Appl. Res. 4 (2009) 127–134. doi:10.1016/j.jveb.2008.10.037.
[33] A.I. Sandem, B.O. Braastad, K.E. Bøe, Eye white may indicate emotional state on a frustration-
contentedness axis in dairy cows, Appl. Anim. Behav. Sci. 79 (2002) 1–10. doi:10.1016/S0168-
1591(02)00029-1.
[34] N. Reefmann, F. Bütikofer Kaszàs, B. Wechsler, L. Gygax, Physiological expression of
emotional reactions in sheep, Physiol. Behav. 98 (2009) 235–241.
doi:10.1016/j.physbeh.2009.05.017.
[35] N. Reefmann, B. Wechsler, L. Gygax, Behavioural and physiological assessment of positive and
negative emotion in sheep, Anim. Behav. 78 (2009) 651–659.
doi:10.1016/j.anbehav.2009.06.015.
[36] R.O. Moe, S.M. Stubsjøen, J. Bohlin, A. Flø, M. Bakken, Peripheral temperature drop in
response to anticipation and consumption of a signaled palatable reward in laying hens (Gallus
domesticus), Physiol. Behav. 106 (2012) 527–533. doi:10.1016/j.physbeh.2012.03.032.
[37] A. Boissy, C. Lee, How assessing relationships between emotions and cognition can improve
farm animal welfare, OIE Rev. Sci. Tech. 33 (2014) 103–110.
[38] E. Verbeek, D. Ferguson, P. Quinquet de Monjour, C. Lee, Generating positive affective states
in sheep: The influence of food rewards and opioid administration, Appl. Anim. Behav. Sci. 154
(2014) 39–47. doi:10.1016/j.applanim.2014.02.011.
[39] H.S. Westerath, L. Gygax, E. Hillmann, Are special feed and being brushed judged as positive
by calves?, Appl. Anim. Behav. Sci. 156 (2014) 12–21. doi:10.1016/j.applanim.2014.04.003.
[40] E.F. Briefer, F. Tettamanti, A.G. McElligott, Emotions in goats: mapping physiological,
behavioural and vocal profiles, Anim. Behav. 99 (2015) 131–143.
doi:10.1016/j.anbehav.2014.11.002.
[41] H.S. Proctor, G. Carder, Nasal temperatures in dairy cows are influenced by positive emotional
state, Physiol. Behav. 138 (2015) 340–344. doi:10.1016/j.physbeh.2014.11.011.
[42] O.H.P. Burman, R.T.S. McGowan, M.T. Mendl, Y. Norling, E.S. Paul, T. Rehn, et al., Using
judgement bias to measure positive affective state in dogs, Appl. Anim. Behav. Sci. 132 (2011)
160–168. doi:10.1016/j.applanim.2011.04.001.
[43] R.T.S. McGowan, T. Rehn, Y. Norling, L.J. Keeling, Positive affect and learning: exploring the
“Eureka Effect” in dogs, Anim. Cogn. 17 (2014) 577–587. doi:10.1007/s10071-013-0688-x.
[44] F. Kuhne, J.C. Hößler, R. Struwe, Behavioral and cardiac responses by dogs to physical human–
dog contact, J. Vet. Behav. Clin. Appl. Res. 9 (2014) 93–97. doi:10.1016/j.jveb.2014.02.006.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
[45] T. Rehn, L. Handlin, K. Uvnäs-Moberg, L.J. Keeling, Dogs’ endocrine and behavioural
responses at reunion are affected by how the human initiates contact, Physiol. Behav. 124 (2014)
45–53. doi:10.1016/j.physbeh.2013.10.009.
[46] L. Gygax, N. Reefmann, T. Pilheden, F. Scholkmann, L. Keeling, Dog behavior but not frontal
brain reaction changes in repeated positive interactions with a human: a non-invasive pilot study
using functional near-infrared spectroscopy (fNIRS), Behav. Brain Res. 281 (2015) 172–176.
doi:10.1016/j.bbr.2014.11.044.
[47] M. Zupan, J. Buskas, J. Altimiras, L.J. Keeling, Assessing positive emotional states in dogs
using heart rate and heart rate variability, Physiol. Behav. 155 (2016) 102–111.
doi:10.1016/j.physbeh.2015.11.027.
[48] K. Nakayama, S. Goto, K. Kuraoka, K. Nakamura, Decrease in nasal temperature of rhesus
monkeys (Macaca mulatta) in negative emotional state, Physiol. Behav. 84 (2005) 783–790.
doi:10.1016/j.physbeh.2005.03.009.
[49] M. Stewart, J.R. Webster, G.A. Verkerk, A.L. Schaefer, J.J. Colyn, K.J. Stafford, Non-invasive
measurement of stress in dairy cows using infrared thermography, Physiol. Behav. 92 (2007)
520–525. doi:10.1016/j.physbeh.2007.04.034.
[50] M. Stewart, K.J. Stafford, S.K. Dowling, a. L. Schaefer, J.R. Webster, Eye temperature and
heart rate variability of calves disbudded with or without local anaesthetic, Physiol. Behav. 93
(2008) 789–797. doi:10.1016/j.physbeh.2007.11.044.
[51] J.L. Edgar, J.C. Lowe, E.S. Paul, C.J. Nicol, Avian maternal response to chick distress, Proc. R.
Soc. B Biol. Sci. 278 (2011) 3129–3134. doi:10.1098/rspb.2010.2701.
[52] K. Kuraoka, K. Nakamura, The use of nasal skin temperature measurements in studying emotion
in macaque monkeys, Physiol. Behav. 102 (2011) 347–355. doi:10.1016/j.physbeh.2010.11.029.
[53] M. Valera, E. Bartolomé, M.J. Sánchez, A. Molina, N. Cook, A. Schaefer, Changes in Eye
Temperature and Stress Assessment in Horses During Show Jumping Competitions, J. Equine
Vet. Sci. 32 (2012) 827–830. doi:10.1016/j.jevs.2012.03.005.
[54] F. Dai, N. Cogi, E.U.L. Heinzl, E. Dalla Costa, E. Canali, M. Minero, Validation of a fear test in
sport horses using infrared thermography, J. Vet. Behav. Clin. Appl. Res. 10 (2015) 128–136.
doi:10.1016/j.jveb.2014.12.001.
[55] L.C. Grandi, H. Ishida, The Physiological Effect of Human Grooming on the Heart Rate and the
Heart Rate Variability of Laboratory Non-Human Primates: A Pilot Study in Male Rhesus
Monkeys, Front. Vet. Sci. 2 (2015) 1–9. doi:10.3389/fvets.2015.00050.
[56] T. Travain, E.S. Colombo, E. Heinzl, D. Bellucci, E. Prato-Previde, P. Valsecchi, Hot dogs:
Thermography in the assessment of stress in dogs (Canis familiaris)—A pilot study, J. Vet.
Behav. Clin. Appl. Res. 10 (2015) 17–23. doi:10.1016/j.jveb.2014.11.003.
[57] M. Zebunke, J. Langbein, G. Manteuffel, B. Puppe, Autonomic reactions indicating positive
affect during acoustic reward learning in domestic pigs, Anim. Behav. 81 (2011) 481–489.
doi:10.1016/j.anbehav.2010.11.023.
[58] M. Zebunke, B. Puppe, J. Langbein, Effects of cognitive enrichment on behavioural and
physiological reactions of pigs, Physiol. Behav. 118 (2013) 70–79.
doi:10.1016/j.physbeh.2013.05.005.
[59] R.O. Moe, J. Nordgreen, A.M. Janczak, B.M. Spruijt, A.J. Zanella, M. Bakken, Trace classical
conditioning as an approach to the study of reward-related behaviour in laying hens: A
methodological study, Appl. Anim. Behav. Sci. 121 (2009) 171–178.
doi:10.1016/j.applanim.2009.10.002.
[60] E. Prato-Previde, S. Marshall-Pescini, Social Looking in the Domestic Dog, in: A. Horowitz
(Ed.), Domest. Dog Cogn. Behav., Springer Berlin Heidelberg, Berlin, 2014: pp. 101–131.
doi:10.1007/978-3-642-53994-7_5.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
[61] P. Valsecchi, E. Prato Previde, P.A. Accorsi, G. Fallani, Development of the attachment bond in
guide dogs, Appl. Anim. Behav. Sci. 123 (2010) 43–50. doi:10.1016/j.applanim.2009.12.012.
[62] M. Gácsi, K. Maros, S. Sernkvist, T. Faragó, Á. Miklósi, Human Analogue Safe Haven Effect of
the Owner: Behavioural and Heart Rate Response to Stressful Social Stimuli in Dogs, PLoS
One. 8 (2013) e58475. doi:10.1371/journal.pone.0058475.
[63] K. Maros, A. Dóka, Á. Miklósi, Behavioural correlation of heart rate changes in family dogs,
Appl. Anim. Behav. Sci. 109 (2008) 329–341. doi:10.1016/j.applanim.2007.03.005.
[64] S. Ioannou, H. Chotard, M. Davila-Ross, No strings attached: physiological monitoring of rhesus
monkeys (Macaca mulatta) with thermal imaging, Front. Behav. Neurosci. 9 (2015) 1–10.
doi:10.3389/fnbeh.2015.00160.
[65] E.F. Hiby, N.J. Rooney, J.W.S. Bradshaw, Behavioural and physiological responses of dogs
entering re-homing kennels, Physiol. Behav. 89 (2006) 385–391.
doi:10.1016/j.physbeh.2006.07.012.
[66] M. Siniscalchi, C. Stipo, A. Quaranta, “Like owner, like dog”: correlation between the owner’s
attachment profile and the owner-dog bond., PLoS One. 8 (2013) e78455.
doi:10.1371/journal.pone.0078455.
[67] M. Stewart, J.R. Webster, A. Schaefer, N.J. Cook, S.L. Scott, Infrared thermography as a non-
invasive tool to study animal welfare, Anim. Welf. 14 (2005) 319–325.
[68] E. Bartolomé, M.J. Sánchez, A. Molina, A.L. Schaefer, I. Cervantes, M. Valera, Using eye
temperature and heart rate for stress assessment in young horses competing in jumping
competitions and its possible influence on sport performance, Animal. 7 (2013) 2044–53.
doi:10.1017/S1751731113001626.
[69] W. Smith, Modern Optical Engineering, 4th ed., McGraw-Hill Education, New York, 2007.
[70] J. Fox, S. Weisberg, An R Companion to Applied Regression, Second, SAGE Publications, Inc.,
Thousand Oaks, 2011.
[71] D. Bates, M. Maechler, B. Bolker, S. Walker, R.H. Bojesen Christensen, H. Singmann, et al.,
lme4: Linear Mixed-Effects Models using “Eigen” and S4, (2015).
[72] A. Kuznetsova, P.B. Brockhoff, R.H. Bojesen Christensen, lmerTest: Tests in Linear Mixed
Effects Models, (2016).
[73] M. Cabanac, S. Aizawa, Fever and tachycardia in a bird (Gallus domesticus) after simple
handling, Physiol. Behav. 69 (2000) 541–545. doi:10.1016/S0031-9384(00)00227-4.
[74] D.M.L. Vianna, P. Carrive, Changes in cutaneous and body temperature during and after
conditioned fear to context in the rat, Eur. J. Neurosci. 21 (2005) 2505–2512.
doi:10.1111/j.1460-9568.2005.04073.x.
[75] V. Engert, A. Merla, J.A. Grant, D. Cardone, A. Tusche, T. Singer, Exploring the use of thermal
infrared imaging in human stress research, PLoS One. 9 (2014) 1–11.
doi:10.1371/journal.pone.0090782.
[76] A. Quaranta, M. Siniscalchi, G. Vallortigara, Asymmetric tail-wagging responses by dogs to
different emotive stimuli, Curr. Biol. 17 (2007) R199–R201.
[77] H.S. Proctor, G. Carder, Measuring positive emotions in cows: Do visible eye whites tell us
anything?, Physiol. Behav. 147 (2015) 1–6. doi:10.1016/j.physbeh.2015.04.011.
[78] F. Kuhne, J.C. Hößler, R. Struwe, Emotions in dogs being petted by a familiar or unfamiliar
person: Validating behavioural indicators of emotional states using heart rate variability, Appl.
Anim. Behav. Sci. 161 (2014) 113–120. doi:10.1016/j.applanim.2014.09.020.
[79] A. Bowman, Scottish SPCA, F.J. Dowell, N.P. Evans, “Four Seasons” in an animal rescue
centre; classical music reduces environmental stress in kennelled dogs, Physiol. Behav. 143
(2015) 70–82. doi:10.1016/j.physbeh.2015.02.035.
[80] M.P. Tulppo, T.H. Mäkikallio, T. Seppänen, R.T. Laukkanen, H. V Huikuri, Vagal modulation
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
of heart rate during exercise: effects of age and physical fitness., Am. J. Physiol. 274 (1998)
H424–H429. doi:10.1097/00005768-199705001-00813.
[81] R. McCraty, M. Atkinson, W. a Tiller, G. Rein, a D. Watkins, The effects of emotions on short-
term power spectrum analysis of heart rate variability, Am. J. Cardiol. 76 (1995) 1089–1093.
doi:S0002914999803099.
[82] J.W.A. Sleigh, J.D. Henderson, Heart-rate-variability and preoperative anxiety, Acta
Anaesthesiol. Scand. 39 (1995) 1059–1061. doi:10.1111/j.1399-6576.1995.tb04229.x.
[83] W.A. Tiller, R. McCraty, M. Atkinson, Cardiac coherence: A new, noninvasive measure of
autonomic nervous system order, Altern. Ther. Health Med. 2 (1996) 52–65.
ACC
EPTE
D M
ANU
SCR
IPT
ACCEPTED MANUSCRIPT
Highlights
4. (Dogs’ emotional responses to palatable food was studied using infrared thermography.) -
optional
5. Eye temperature and HR increase while receiving treats, no clear pattern of HRV emerges.
6. Tail wagging could be a useful indicator of positive emotional state.
7. Eye temperature can assess the arousal dimension of dogs’ emotions in a positive situation.
Keywords: Dog; Positive emotion; Behavior; Infrared thermography; Heart Rate; Heart Rate
Variability